Upload
kogyi-kyaw-kwo
View
221
Download
0
Embed Size (px)
Citation preview
7/28/2019 Types of Research Methods.pdf
1/2
Types of Research Methods
AdaptedfromEdvantiaSBRRatingforTechnicalAssistanceProgramsandServicesform(2007)andCarterMcNamaraOverviewofMethodstoCollectInformationhandout(1998)
2008 by The SERVE Center at the University of North Carolina at Greensboro. All rights reserved. www.serve.org
Evidence of
effectiveness
Research
Method
This is This works best for these
kinds of questions
This doesnt work well
for these kinds of
questions
Additional Things
to Note
LOW Descriptive-
Qualitative
(Ethnography/
Case Study)
Detailed descriptions of
specific situation(s) using
interviews, observations,
document review
You describe things asthey are.
How do people implement
this program?
What challenges do people
face?What are peoples
perceptions?
Did the program cause
any changes in
participants outcomes?
Descriptive-
Quantitative
Numerical descriptions
(frequency, average)
You measure things as
they are.
How many people are
participating in this program?
What are the characteristicsof people in this program?
How well did participants in
this program do?
Did the program cause
any changes in
participants outcomes?Why did the program
work this way?
Correlational/Regression
Analyses
Quantitative analyses of
the strength ofrelationships between twoor more variables (e.g., are
teacher qualifications
correlated with student
achievement?)
What is the relationship
between various school orclassroom context factors andstudent achievement?
Is the extent of
implementation of a program
across sites correlated with
better outcomes?
Did the program cause
any changes inparticipants outcomes?
Look for words such as,
more likely than, lesslikely than, associatedwith, related to, and
correlated with.
Quasi-
experimental
Comparing a group that
gets a particularintervention with another
group that is similar in
characteristics but did not
receive the intervention
norandom assignment
used
Did the program cause any
significant differences inparticipants outcomes as
compared to non-participants
with similar characteristics
who did not receive theintervention?
How are people
implementing theprogram?
Why did the program get
the results it did?
Look for the phrase
compared with.
Look for results that are
both statistically significant
and meaningful.
NOTE: Did the study test
the equivalence of treatment
and control groups prior tothe intervention?
7/28/2019 Types of Research Methods.pdf
2/2
Types of Research Methods
AdaptedfromEdvantiaSBRRatingforTechnicalAssistanceProgramsandServicesform(2007)andCarterMcNamaraOverviewofMethodstoCollectInformationhandout(1998)
2008 by The SERVE Center at the University of North Carolina at Greensboro. All rights reserved. www.serve.org
Evidence of
effectiveness
Research
Method
This is This works best for these
kinds of questions
This doesnt work well
for these kinds of
questions
Additional Things
to Note
HIGH
Experimental
Using random
assignment to assignparticipants to an
experimental or treatment
group and a control orcomparison group (e.g.,
one receives theintervention and one doesnot)
Did the program cause any
significant differences inparticipants outcomes as
compared to the control
groups outcomes?
How are people
implementing theprogram?
Look for words such as,
causes or leads to.
Look for results that are
both statistically significantand meaningful.
NOTE: The interventionshould be clearly defined
so that you know what it
was designed to entail, and
to what extent it was
implemented in the study.Also look for information
on the experience of the
control group.
Meta-analysis
Synthesis of results from
multiple studies to
determine the averageimpact of a similar
intervention across the
studies
Over all studies conducted on
a particular intervention or
strategy, what can be saidabout the direction or
strengths of the impacts?
What does the totality of
research studies say about theeffectiveness of a program?
How are people
implementing the
program?What are peoples
perceptions?
Look for selection criteria
used to include studies and
look for measures of effectsize.
Look for differences in
results among the studies.Do some studies show
positive results while others
show negative or do allstudies show positive
results?